The Arts and Humanities are among the early adopters of computer technology. In the late 1940s, the linguist and Jesuit Father Roberto Busa S.J. convinced Thomas J Watson and IBM that the exploration of Thomas Aquinas' works could be more efficient using new digital methods. The index Thomisticus and the Corpus Thomisticum - a digital lemmatized collection of all works of Thomas Aquinas have gone through all phases of technological development - via punched cards, magnetic tapes and the WWW - and are still accessible today. Since then, a multitude of artifacts and resources have been digitized, so that by now although only a fraction of the entire cultural heritage is digitally available, a critical mass for the development of digital research methods nevertheless exists. The talk will show exemplary which strategies Digital Humanities use to generate, interlink and explore knowledge from historical sources.

There are more than 10 millions astronomical photographic negatives (glass plates and planfils/films) worldwide. Most of them represent stellar fields and solar images, but there are also numerous negatives with other celestial objects. There is large variety of these negatives (wide field and narrow field direct images, multiple images, low dispersion and high dispersion spectra etc.). I will give review and discussion of these databases including recent progress in their digitization and scienfic use.

We all know that Astronomy has a wonderful, rich, diverse but distributed heritage of photographic plates, counting into the millions. Most are well ordered, and in a reasonable (though often rather questionable) state of preservation. As such, they are the envy of many other natural sciences. But Astronomy's heritage data are generally unstaffed, lack routine monitoring and care, and lack any future if they stay as they are. Some efforts are now being made, and some of the collections are being, or have been, digitized in some manner; Harvard is making good progress with digitizing the biggest of all collections with a purpose-built machine. But there are many other collections, in various countries, with no future, and if nothing is done on a global scale we risk losing a great deal of unrepeatable science. Positive action on that global scale is needed as a matter of urgency, before the expert knowledge about handling and interpreting photographic observations is no longer available to us. This paper will describe what is being discussed to those ends, and seeks space for a plenary discussion as to what is feasible.

Direct plates archives may be used to remeasure stars which were the target of the observations or to look at other stars discovered to be interesting only at a later time. I present here two cases: a search for Long Period Variable (LPV) stars on IR plates, and a search for past outbursts of a recently discovered cataclismic variable. Long period red variables (LPV) have been searched in a collection of Schmidt plates of the Asiago Observatory taken in the I band, centered on Gamma Cas and covering the years 1967-1975. These plates were not used by the original investigators. Digitization was made at the Perugia University and analysis was performed with the PyPlate software developed for the APPLAUSE project. Twentyone variables were found, nearly doubling the number of presently known variables in the field. Ten stars have Mira light curves, two of them with marked double period. Spectroscopic follow-up gave a large fraction of Carbon stars (7/21) and of S-type stars (6/21). Blue plates of the same field were also taken in pair with the IR ones and I have used them to explore the past behaviour of ASASSN-18aan, a CV variable recently discovered robotically. Thre past flares were identified and a possible recurrence time scale of about one year was found.

Direct plates archives may be used to remeasure stars which were the target of the observations or to look at other stars discovered to be interesting only at a later time. I present here two cases: a search for Long Period Variable (LPV) stars on IR plates, and a search for past outbursts of a recently discovered cataclismic variable. Long period red variables (LPV) have been searched in a collection of Schmidt plates of the Asiago Observatory taken in the I band, centered on Gamma Cas and covering the years 1967-1975. These plates were not used by the original investigators. Digitization was made at the Perugia University and analysis was performed with the PyPlate software developed for the APPLAUSE project. Twentyone variables were found, nearly doubling the number of presently known variables in the field. Ten stars have Mira light curves, two of them with marked double period. Spectroscopic follow-up gave a large fraction of Carbon stars (7/21) and of S-type stars (6/21). Blue plates of the same field were also taken in pair with the IR ones and I have used them to explore the past behaviour of ASASSN-18aan, a CV variable recently discovered robotically. Three past flares were identified and a possible recurrence time scale of about one year was found.

Sonneberg Observatory run photographic sky surveys (Sky Patrol and Field Patrol) from 1923 through 2010, the plates of which have mostly been digitized. The goal of continuation of the surveys by digital means has gradually been realized by fish-eye cameras and small telescopes monitoring of selected fields. The talk provides a brief overview on current scanning, observing and analysing activities.

From the beginning of the photographic observations by Küstner in Bonn at 1900 it was planned that future astronomers may use these data for the determination of proper motions. This idea was realised by astronomers of Bonn starting in 1981. Their efforts resulted mainly in determinations of proper motions of stars in open and globular clusters. Moreover, astronomers from Hoher List observatory took plates for the search of variable stars.

The Dr. Remeis-Sternwarte in Bamberg, the Leibniz-Institut für Astrophysik, Potsdam and the Hamburger Sternwarte host important astronomical plate archives. Astronomers from those institutions teamed up to digitize their archives, calibrate and integrate them into the publically available database APPLAUSE. In an international collaboration also the plate archive of the Tartu observatory was included. Here we briefly describe the instrumentations used and the history of the plate archives of the collaboration.

The star HBHA 1704-05 was originally classified as an emission-line star (Kohoutek and Wehmeyer, 1999). On the basis of The All Sky Automated Survey for SuperNovae (ASAS-SN), the object was cataloged in the VSX as an semi-regular variable with a periodicity around 418 days. At the beginning of August 2018, the ASAS-SN survey indicated a rapid brightening of the star. Following high-cadence photometric and spectroscopic observations clarified the nature of HBHA 1704-05 as a symbiotic star in outburst. Monitoring the object within the ASAS-SN survey for around 3.5 years revealed a wave-like variation with the amplitude DV ~ 0.7 mag and a period of ~ 495 days. Our close inspection of the Moscow's photographic plates archive revealed another 2-mag outburst of HBHA 1704-05 lasting from 1968 to around 1990, during which a regular variation with the same period and phase, but amplitude of D(m__pg) ~ 1 mag, emerged. These properties support the nature of HBHA 1704-05 as a symbiotic binary with the orbital period of ~ 495 days. Close similarities with other symbiotic systems are briefly discussed.

Cataclysmic variables and their relatives (low-mass X-ray binaries) are very active objects. We summarize the dominant features of their activity. We show that monitoring by various types of instruments is necessary to detect the events (e.g. outbursts, transitions between the high and low states) which are often unpredictable. Histograms of brightness provide us with the properties of activity of various types of such objects. Photographic monitoring spanning for about a century enables to study the properties of the light curves. We also show how the photographic light curves can be combined with the CCD and visual observations. The long monitoring also shows that the activity of some objects dramatically changed during several decades. We also briefly summarize the astrophysical interpretations of the long-term activity.

Photographic emulsion is a non-linear light detector which makes it difficult to perform photometry with digitized plates using conventional software. Inexpensive flatbed scanners (often used to digitize plates) provide image quality acceptable for photometry, but introduce complex image distortions (notably, the hacksaw pattern) that complicate astrometry and source identification. VaST is a photometry package designed from the ground up to handle such images. It relies on SExtractor for source detection/photometry, matches source lists derived from plates exposed at different epochs and cross-calibrates their magnitude scales to construct lightcurves of all the detected objects. An array of statistical methods can be applied to these lightcurves to identify variable objects. Accurate celestial positions of variable objects are measured by using nearby sources to compute local corrections to the approximate plate solution obtained with Astrometry.net. VaST is used mostly with the Moscow collection plates, but is intended to be a general-purpose tool for lightcurve extraction from a series of digitized photographic images.

Digitizing an archive of photographic plates yields a set of image files and accompanying metadata files. To handle and process these data efficiently, we have developed a Python package PyPlate. PyPlate provides methods to read data from CSV and FITS files, to make calculations with various observation timestamps, to create consistent and easily readable FITS headers, to write plate metadata into a database, to extract sources from plate images, to carry out astrometric and photometric calibration on the extracted sources, and to output the source data to a database or files on disk. We will show how the PyPlate software was used for building the APPLAUSE database and discuss how it can be applied elsewhere.

Digital scans of photographic plates are more complicated to handle in comparison to the CCD frames. Due to their fragile nature and age they are subject to several effects that act as a source of false-positive detections. Among these effects are surface dust speckles, scratches, emulsion damage, and writing traces. Besides those the optical systems used to acquire the images come with their share of aberrations which make the removal of false-positives harder. The APPLAUSE database already consists of a few billion detected sources, many of which are not of astrophysical nature. Their detection and / or removal is very important for practical purposes but it is obviously not a trivial task. We will present a convolutional neural network based pipeline that efficiently and quickly weeds out the false-positive detections and is also capable of assigning false-positive probability to individual sources in less clear-cut cases. Our plan is to use its output in the next APPLAUSE data release.

In this three-part session we (T. Tuvikene, H. Enke) will first present the concept of the archive, describe its building blocks and features.